Although obesity, diet, and physical activity have all been linked to incidence of type 2 diabetes, there are many unanswered questions about the interactions of these factors on risk of this disease. Furthermore, there is a lack of informative data on the relation of diet and physical activity to risk of type 2 diabetes in black women. Data from black women are sorely needed. US black women have a considerably higher incidence of type 2 diabetes than white women and the difference does not appear to be explained by different prevalences of risk factors. At every level of body mass index black women have a greater incidence than white women, and the relative increase is even higher among the most obese women. Whether this reflects metabolic differences or differences in other environmental factors is unknown. Diet and physical activity patterns are culturally influenced and differ between black women and the white populations that have been extensively studied. Examining these factors in a large cohort of black women will permit the evaluation of dietary and activity patterns specific to US black women and may yield more useful data for intervention programs. Data from black women may also have a greater impact on motivating black women to modify their behaviors. The Black Women's Health Study (BWHS) is an ongoing cohort study of 59,000 US black women enrolled in 1995 and followed by biennial mailed questionnaire. By the end of the proposed study period there will be an estimated 3150 verified incident cases of type 2 diabetes in the cohort, the majority of which will have occurred in participants who are overweight or obese. We propose to use BWHS data to examine interactions of diet, physical activity and obesity in relation to incidence of type 2 diabetes. Specific lifestyle factors to be studied include types of fat in the diet, glycemic load of food intake, alcohol intake, coffee intake, walking, moderate activity, vigorous physical activity, and number of hours watching TV. In addition, under a novel aim, census bureau data will be linked with questionnaire data to permit analyses of neighborhood characteristics in relation to obesity and type 2 diabetes. Because the follow-up and data collection are funded by another grant, the proposed large study of modifiable risk factors for type 2 diabetes can be carried out at a relatively modest cost.